Home > Uncategorized > Hunting for causes (wonkish)

Hunting for causes (wonkish)

from Lars Syll

Causality and CorrelationThere are three fundamental differences between statistical and causal assumptions. First, statistical assumptions, even untested, are testable in principle, given sufficiently large sample and sufficiently fine measurements. Causal assumptions, in contrast, cannot be verified even in principle, unless one resorts to experimental control. This difference is especially accentuated in Bayesian analysis. Though the priors that Bayesians commonly assign to statistical parameters are untested quantities, the sensitivity to these priors tends to diminish with increasing sample size. In contrast, sensitivity to priors of causal parameters … remains non-zero regardless of (non-experimental) sample size.

Second, statistical assumptions can be expressed in the familiar language of probability calculus, and thus assume an aura of scholarship and scientific respectability. Causal assumptions, as we have seen before, are deprived of that honor, and thus become immediate suspect of informal, anecdotal or metaphysical thinking. Again, this difference becomes illuminated among Bayesians, who are accustomed to accepting untested, judgmental assumptions, and should therefore invite causal assumptions with open arms—they don’t. A Bayesian is prepared to accept an expert’s judgment, however esoteric and untestable, so long as the judgment is wrapped in the safety blanket of a probability expression. Bayesians turn extremely suspicious when that same judgment is cast in plain English, as in “mud does not cause rain” …

The third resistance to causal (vis-a-vis statistical) assumptions stems from their intimidating clarity. Assumptions about abstract properties of density functions or about conditional independencies among variables are, cognitively speaking, rather opaque, hence they tend to be forgiven, rather than debated. In contrast, assumptions about how variables cause one another are shockingly transparent, and tend therefore to invite counter-arguments and counter-hypotheses.

Judea Pearl

Pearl’s seminal contributions to this research field is well-known and indisputable. But on the ‘taming’ and ‘resolve’ of the issues, yours truly however has to admit that — under the influence of especially David Freedman and Nancy Cartwright — he still has some doubts on the reach, especially in terms of realism and relevance, of Pearl’s ‘do-calculus solutions’ for social sciences in general and economics in specific (see hereherehere and here). The distinction between the causal — ‘interventionist’ — E[Y|do(X)] and the more traditional statistical — ‘conditional expectationist’ — E[Y|X] is crucial, but Pearl and his associates, although they have fully explained why the first is so important, have to convince us that it (in a relevant way) can be exported from ‘engineer’ contexts where it arguably easily and universally apply, to socio-economic contexts where ‘surgery’, ‘hypothetical minimal interventions’, ‘manipulativity’, ‘faithfulness’, ‘stability’, and ‘modularity’ are not perhaps so universally at hand.

CAUSES on Twitter: ""Right now, whole genome testing is most useful for  helping unravel the mystery for parents of children with rare disorders; it  can provide an answer about the cause, butWhat capacity a treatment has to contribute to an effect for an individual depends on the underlying structures – physiological, material, psychological, cultural and economic – that makes some causal pathways possible for that individual and some not, some likely and some unlikely. This is a well recognised problem when it comes to making inferences from model organisms to people. But it is equally a problem in making inferences from one person to another or from one population to another. Yet in these latter cases it is too often downplayed. When the problem is explicitly noted, it is often addressed by treating the underlying structures as moderators in the potential outcomes equation: give a name to a structure-type – men/women, old/young, poor/well off, from a particular ethnic background, member of a particular religious or cultural group, urban/rural, etc. Then introduce a yes-no moderator variable for it. Formally this can be done, and sometimes it works well enough. But giving a name to a structure type does nothing towards telling us what the details of the structure are that matter nor how to identify them. In particular, the usual methods for hunting moderator variables, like subgroup analysis, are of little help in uncovering what the aspects of a structure are that afford the causal pathways of interest. Getting a grip on what structures support similar causal pathways is central to using results from one place as evidence about another, and a casual treatment of them is likely to lead to mistaken inferences. The methodology for how to go about this is under developed, or at best under articulated, in EBM, possibly because it cannot be well done with familiar statistical methods and the ways we use to do it are not manualizable. It may be that medicine has fewer worries here than do social science and social policy, due to the relative stability of biological structures and disease processes. But this is no excuse for undefended presumptions about structural similarity.

Nancy Cartwright

  1. Ken Zimmerman
    May 5, 2021 at 6:45 am

    When it comes down to anthropological praxis, discreteness and continuity are both a matter of professional judgment and a product of the partic- ular problems and issues on which one is working. Both discreteness and continuity are arbitrary in an absolute sense, and, therefore, each theorist needs to draw upon the literature, their own experience, the views of their informants, and their individual creativity to build an appropriate classificatory model of the units of causality and a suitable theory of dynamism. Then the theorist must merge these two models together in the warp and woof of an argument which justifies the boundaries and distinctions one has drawn in time and space, as well as the causal scheme one has used to explain the continuous process of change. Then, as Sally Falk Moore quips, “it is customary to argue for their inherency in the research material” (pers. comm. 1991).

    A less painful approach to causality than devising empirical tests of the so-called “true nature” of causality is to interview informants in the field as to their own ideas about causality. In speaking of mana, orenda, or the totemic principle in “The Elementary Forms of Religious Life,” Durkheim was alluding to such an approach (1915:406). This approach seems to have ironic implications for Tambiah’s scheme in “Magic, Science, Religion, and the Scope of Rationality” in which Tambiah suggests that one characteristic difference between causality and participation is that causality “explains,” whereas participation “expresses” (1990:9). Yet, although to the anthropological researcher another culture’s emic conception of causality may “express” that culture’s members’ experience of participating in the world, to many members of that culture, their conception of causality may “explain” the world. In addition, emic and etic conceptions of causality sometimes coincide, whether the anthropologist is an outsider or an insider to the culture.

    Looking more closely at Sahlins” critique of causation, I would like to say that I find Sahlins’ blunt dismissal of mutual causation of nature and culture especially disturbing. At the very least, like yin and yang in Chinese philosophy, the categories of nature and culture depend on each other for their distinctive existence. But Sahlins seems to think that mutual causation of nature and culture would necessarily imply fatalism. I argue that this is not necessarily the case. Although I see how Sahlins might be turned off by Bateson’s rendition of mutual causation of nature and culture (1972), I do not see why we have to limit our consideration to Bateson’s particular ideas about the notion. I share Sahlins’ skepticism over Bateson’s “cybernetic alternative” which holds that culture and nature are mutually determining because they are part of a total system of “immanent mind” whose process involves reciprocal interactions between nature and culture (Sahlins 1976: 90 f.). Yet, there are other alternatives. Were social anthropologists to focus on developing theories of context-dependent probabilistic causality (John Dupre 1984; Tambiah 1990), then we could entertain the likelihood that sometimes nature causes culture, sometimes culture causes nature, and sometimes nature and culture engage in mutual causation (Bunge 1979: 62) and sometimes the two interact in noncausal, processual ways (Addis 1984; Achinstein 1984) all dependent on the complex web of contingencies in a particular space in time. At this juncture, I must admit that despite my being intrigued with the idea of context-dependent cause, while this notion solves some theoretical problems, at the same time it raises others, such as the problem of how to decide what to call context and circumstance and what to call cause. It seems as though this could be an arbitrary decision. Why should I not decide to call “context” and “circumstance” causal factors as well if they are indeed highly crucial to the production of the effect? Of course, this harkens back to the debate about sufficient and necessary causes and the frequent difficulties one encounters in distinguishing the two.

    In “The Domestication of the Savage Mind,” Jack Goody puts forth a view alluding to the need for greater creativity on the part of anthropologists in the ascription of causal explanations.

  2. May 7, 2021 at 12:23 pm

    It seems to me this discussion needs to start with ontology, i.e. what a cause IS. In Newton’s flat earth understanding of space a cause is a property of a pre-existing particle or “atom”, whereas in Einstein’s big bang spherical understanding of space, in which matter and energy are interchangeable, the cause is originally an energetic PROCESS whereby some of the energy became localised by circulation to form a spray of Newtonian particles which themselves became bound together as atoms, molecules and eventually us by energetic causes, including that already captured in subatomic particles.

    Nancy says “It may be that [evidence based] medicine has fewer worries here than do social science and social policy, due to the relative stability of biological structures and disease processes. But this is no excuse for undefended presumptions about structural similarity”.

    Indeed; but the point is that the structures are made up of particles, which have been discovered, not caused, by technologists, while the periodic table in chemistry and the genetic, vascular and neurological structures underlying medicine are hardly presumed or undefended.

    Ken asks “Why should I not decide to call “context” and “circumstance” causal factors as well if they are indeed highly crucial to the production of the effect?” I am trying to say there are two views of cause which need to be distinguished: dynamic energy (often thought of as static power, force or mass) and the relatively static material structures which direct the energy.Thus in the context of our monetary economy it is predictable that I will walk down a road to a bank but not when I will be moved to do so. That is my Einsteinian interpretation; it seems to me Ken is still offering a Newtonian one. As I see it, “context” and “circumstances” come into the detailing of the effect, not causing it.

    • Ken Zimmerman
      May 7, 2021 at 11:15 pm

      Dave, my comment is part of a langer article from several years back. It is written as an anthropology of Sahlins’ review of causation in anthropological works. I try to make clear that for anthropologists anthropology (creation of human lives) is ontology. Humans make their being in creating their cultures and societies. Following this work is the role of anthropologists. Leading sometimes to anthropologists being perceived as oddballs by those they observe and on whom they report.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.